Upper gastrointestinal malignancies (esophageal, gastric and pancreatic) are one of the leading causes of cancer deaths among Americans largely due to its insidious nature leading to diagnosis at late stages. The clinical implementatio of ultrathin upper endoscopes has the potential for providing a major modality for cancer screening in that it is much more comfortable than conventional larger caliber endoscopes thus allowing patients to forego sedation with the consequent change in complications, cost and convenience. However, the promise has heretofore not been realized largely because of the failure of upper endoscopy to detect pancreatic cancer (PC), whose incidence dwarfs esophageal and gastric cancers. Our multi-disciplinary group of biomedical engineers, cancer biomarkers experts, gastroenterologists has focused on using a novel technology, low coherence enhanced backscattering (LEBS) to detect the ultra-structural and microvascular consequences of the genetic/epigenetic/physiological alterations in field carcinogenesis. While our prior work has largely focused on the colon (where the condemned mucosa is the clinical imperative behind assessment for synchronous/metachronous adenomas), our data have shown that the analysis of histologically and endoscopically normal duodenal peri-ampullary mucosa could predict PC (consistent with other reports on genetic/epigenetic events). Our preliminary data with a LEBS fiber-optic probe requiring a large working channel shows strong diagnostic performance. For this strategy to be clinically viable, however, the probe has to be miniaturized so that it can be delivered through ultrathin endoscopes. Therefore, we propose to develop a novel miniature sub-diffusion radiative transport fiber-array probe to be compatible with the 2 mm accessory channel of an ultrathin endoscope. We will assess clinical value in two scenarios: PC screening and cystic neoplasm management. With screening we will perform a case-control study with an independent validation set to test the potential of detecting the ultrastructural and microvascular markers of field carcinogenesis in the duodenal peri-ampullary mucosa via the probe for PC screening. For cystic neoplasm management, we will assess the value of duodenal interrogation for cyst diagnosis and prognostication of cysts with intermediate malignancy risk (IPMNs), both clinically vexing issues. By bridging ultrathin endoscopy with optical detection of PC field carcinogenesis in the duodenal periampullary mucosa, we propose to overcome what is arguably the major barrier thus far preventing upper GI cancer screening in population. The test could be performed comfortably in the office setting while minimizing cost, patient inconvenience and complications. This may also have applications for surveillance of patients at higher risk including those with IPMNs. Finally, this may herald another avenue in biophotonics research focusing on risk stratification as opposed to optical biopsy or dysplasia detection.